On the Science of Second Chances | Jennifer Doleac

0:00Once you're actually in that room, it was just amazing to see, especially all these lawmakers from across the political spectrum, far left to far right, all in the same room, because they care about the same issue and they're trying to, you know, do something with domestic violence, or they're trying to do something about all the money we're spending on prisons, right? Or whatever their challenge is in their state, and kind of realizing like, wait, I might never have a conversation with you if you are, you know, in the same legislature as me, because you'd be on the other side, but it turns out we both are trying to solve the same problem in good faith. And so how can we work together to do that? And there were so many examples of these bipartisan efforts that would come out of that event for exactly that reason. It kind of gave people a chance behind closed doors to realize they actually had more in common than they thought, and that was just so cool to see. When criminal justice expert and Stanford trained economist, Dr. Jennifer Dolyak thinks about reform?

1:00She's not just hopeful. She's optimistic that second chances are possible for the justice involved population and the system as a whole. In her eye opening new book, The Science of Second Chances, she reveals her powerful approach to reducing crime and incarceration. Today, we'll take a deep dive into some of the cutting edge economic research and real world experiments she's conducted and called upon in her work, together exploring the possibility of approaching big, complex problems as scientists in a lab do, carefully testing different approaches and following the data to maximize impact. Jennifer's work points to an innovative path forward that harnesses practical, achievable solutions. She invites us to consider the possibility

2:02that we can have both public safety and a smaller, less intrusive justice system without waiting for big structural reforms that might never come, and that creating the society that we ultimately want and need may actually come from shifting focus from the enormity of the problem to the power of small evidence-based changes. For more on Jennifer, her new book, Podcast, and other offerings, please visit Jennifer Dolyak.com and for her sub-stack, head to probablecazation.substack.com. I'm Joshua Steinfeld, and this is the courageous life.

3:05Well, Jennifer, welcome to the show. Hi, thank you for having me. It is such a pleasure to be with you. I just finished reading your book, The Science of Second Chances, a revolution in criminal justice. And for me, it's one of those books that I think this is actually like a key takeaway for me in reading the book, which is sometimes well-intentioned policies or processes or systems or rules we want to put in place to try and help other people or help better society or whatever it might be. Actually, can fall flat, they can even backfire, they can even do the worst, which is cause harm to the people who are trying to help. And you really uncovered using rigorous research and all these examples, both the research that you've conducted, but also a lot of other people have conducted as well about what works and what's effective. So I'm sitting here reading this book and it's just eye-opening to me.

4:05It was just really beautiful to say and one of the things too that we can talk about a bit later is sometimes like the idea of less is more. To like even small changes, small tweaks that we make sort of, oh, that's not gonna work to drop in the bucket or something. Actually, kind of an unexpectedly large impact. So thank you, I feel like your work is so incredible and it's so important and needed right now. Thank you for that. I'm an economist by training and I think economists, you know, we're known as the dismal scientists and we're definitely good at or becoming used to bringing bad news to policy conversations. We're often the ones in the room saying that might not work or that didn't work, but you know, when I try to also convey, I hope it came through in the book, but when I try to convey in my conversations with policymakers and practitioners too, is just that I am extremely optimistic about this policy space, most policy spaces, I think there are solutions out there. And really, you know, the reason,

5:06I think it's important to bring the bad news to those conversations when something doesn't work, is that it's just slowing us down in our efforts to find the things that will work and that will make people's lives better. So, you know, I think people often hear me talk about the need for research and evidence as if I'm just trying to slow things down. And actually, it's the opposite. I want to go faster and I want to find solutions faster. And I think it's possible. Yeah, as I was reading the book, I was reminded of a lot of things, but I was reminded of a past guest named Ronda McGee and Ronda McGee as a lawyer at the University of San Francisco. She teaches law. And she's also a leader in the mindfulness space. She kind of combines mindfulness into social justice work. And one of the things that she shared as we were having this conversation is kind of this counter-cultural mindset that she has when she engages in that type of work, which is she goes, I sort of train myself to fail. Like, I sort of know or expect, I'm gonna mess up when I'm having conversations across difference

6:07and we're doing social justice work. Like, I'm gonna say the wrong thing. I'm gonna disappoint somebody. I'm gonna piss somebody off. Like, it's just part of it. And I thought that mindset, it is counter-cultural, I think, but also like, so helpful. And to me, like just pulling something out of the book, you point to something similar, which is sort of like an expectation that a lot of things maybe won't work, but we will find something that will. So, I don't know, I just, I love that. I love that connection. Yeah, one of the things I find myself saying over and over is our goal should be to fail fast, rather than for tend we won't fail, because all the problems that still exist in the world that we're trying so hard to solve, there's still a problem because they're hard to solve, right? It was so obvious and so easy to fix. We would have done it already. And so we should be assuming, at this point, most things we're trying probably won't work. The first thing we try isn't gonna work, or at least won't work as well as it could, right? So our goal should be to fail fast, iterate, try again.

7:07And I think that's a common mindset in like the tech and startup world, right? Like Silicon Valley, those founders love failing fast. It's very uncommon in the policy space. No one wants to admit that they might have gotten it wrong. And I think that's really to our detriment. Yeah, I mean, just to one of many useful insights that came out of reading the book for me anyway. So, you know, like before we dive further in, I'd love to kind of back up and invite you to share some of your story. And the way that I always invite people to do this because this is the courageous life is to really think about whether there was a time where you faced some sort of adversity or challenge. And this could really honestly be anywhere in life. It could be your childhood, it could be the start of your career, anywhere along the way. That you feel sort of put you on your path or influence your trajectory, largely what you're doing professionally in the world today. So does anything like that come to mind for you just as a way in? Yeah, you know, it's like, I think especially talking about the issues

8:07or going to be talking about crime and criminal justice policy, it feels ridiculous to say like anything I have faced is think of it as a hardship because I've been pretty lucky through my life and career. But, you know, the first thing that comes to mind is, it was a professor for many years. So it was a, when to graduate school, got a PhD, was an economics professor for many years. And, you know, didn't necessarily start out saying that was the path I wanted to be on was really interested in policy, but eventually became, you know, was very in that world, a very wed to it. And then there was a point around, I guess it was 2022, where I became very involved in conversations around sexual harassment in the economics profession and academia were broadly. And luckily, I have not been evicted myself, but I do have friends that have been. And it turns out I just have a really deep, seated sense of justice. I've learned about myself, like I just really, I find it completely unacceptable to see harm being done, especially people in positions of power get away with no consequences.

9:09That just like really, I cannot stand it. It's a theory. It's absolutely intimidating. And I just, I don't understand. I think there are a lot of people in academia who are very good at just going back in their offices and shutting the door and getting back to work. And I am just, I've never been that kind of person for better and worse than I think. But, you know, I became very, you know, relative to a lot of other academics, I'm good at being allowed on the internet. I joke to people. And so, you know, there were a few really high profile incidents in 2022 where there were no consequences for the men in that case involved. And I started just raising a lot of awareness, say, and being allowed on the internet about it and calling some people out for actions that I thought were really unhelpful in the field. And really just trying to get people to take it seriously. And it turns out that really, you know, a lot of good came from that. A lot of people were very happy that I was taking those positions. And a lot of people were really upset that I was taking those positions, including people in my own department and university.

10:10And so, kind of wound up going down this road where, you know, I definitely faced professional retaliation internally at the university I was working at. And I had mentors that came to me and said, you know, I had a sabbatical coming up saying, you know, I had tenure at the time. So I figured that protected me. And I had mentors that came and said, you should expect that when you come back from your sabbatical, your colleagues will stop letting you buy out your teaching, give you lots of new course crafts, basically make you want to quit this job. And I realized at that point that tenure really had no value. You know, you kind of think you work really hard as an academic to get tenure that guarantees you your job as a professor and researcher, you know, regardless of what your public stances might be. And it kind of realized in an instant that was not the case. And so that led me to leave academia. I mean, within a month, I was like, all right, let's see what my other options are. And so I, you know, suddenly became open to having external conversations. And then that led me almost immediately to this job

11:10at Arnold Ventures, or I was definitely in the criminal justice team for the last three years. And it's been a great ride. And so it all ended well, but it definitely was a really stressful few months. And it's a sort of thing that I really don't want other people to be going through. in your last few weeks of your career, I I even know you were all confused about that. Now, life is a very IFU, it gets different in the beginning of the life, and I think trying to help while I was in general 對不 I had a選擇檁, I feel like that motivates me to do that but anyway, I like to ask to arrange and Earn some �able people to stay comfortable. Trying to pull everyone out because I didn't go back until my close up year and I had the longest break. I just interviewed this woman named Dr. Sunita Sa, and she wrote a book called DeFi, and the subtext is how to speak up when it matters. And you just really reminded me of that because you spoke up when it mattered, you know, and how that's not always easy to do. And I guess my own curiosity,

12:13she points to a couple of things that she's seen in her research about this. And one of them is the kind of can diagnose or do this assessment in a moment for these two dimensions. One is safety, and one is impact. Is it safe to speak up what I be in physical danger? I'm going to be harmed if I speak up now. We hurt or whatever it might be. And is it the right time in terms of like, can I make an impact here? Recognizing that sometimes we need to speak up, and it's not safe. And it's just that's just the reality, right? So I guess pointing to those two things, I'm just curious if there's anything that you learned, you know, looking back on this experience that if people are out there and I recognize this is a big and nuanced question, but just from like your own experience, for someone who's wanting to speak up about something that matters. Is there anything that you can point to kind of about how to do that skillfully or any reflections you have from this experience? You know, I think on net, and there have been other instances too, where, you know, I feel like I

13:15talk about research, people don't like, and people come after me on the internet, and a whole bunch of stuff like that. I've certainly been through that ringer several times. I think on net through each of those experiences, I've come out ahead professionally. So in terms of like what I care about, right? In terms of like making clear that I have, you know, what my personal integrity is and that I'm willing to kind of speak truth to power. And, you know, I think there are a lot of people become concerned about being canceled at the SOTIS fake, and my own experience is that if you pick the right fights, right? If you stand up in those right in those key moments when it really matters, it's hard for me to think of examples of people that didn't come out on top in the end. And that's certainly been my experience. And, you know, it's hard as it is in those moments. It winds up bringing me to a better place. The fact that I left academia and am now in this new world and these kind of on this new career path, I am a hundred times happier than I was, and it pushed me to make a leap

14:17maybe should have made years earlier in some ways. And so yeah, I think there are a lot of, I know a lot of people who mean very well who are incredibly risk-vers and worry that it's only going to hurt them professionally. They stand up and, you know, point out something that isn't right. And I think the bigger risk is when people stay silent. I wish more people were brave, more silence as a silence as a big cost to it as well. Yeah. Thank you so much for sharing some of that story. And I guess since we're talking about speaking up and fighting that actually kind of leads us into like, I kind of wanted to explore this conversation in two parts with you. So kind of the second part of the latter part of the conversation will be more about like what works and what you've learned about that and different things you've tried. But the first part, I wanted to kind of introduce it this way. So earlier this year, I interviewed Matt Kaplan. Matt Kaplan is a science correspondent for the economist. And you might be familiar with him or his work, I don't know.

15:20But he came out with this book called, I told you so, scientists who are ridiculed, imprisoned, and exiled for being right. And he told all these stories of all these scientists who were making these discoveries and pushing great work forward and they were breakthroughs and different things. But people imprisoned them or exiled them or tarnished their reputation or destroyed their career or whatever it might be. So there are all these consequences. And I'm reading your book. And I think there's this moment and you can correct or edit anything that I'm not getting right here. But there was this moment where you were talking about employment. And as we're applying to jobs, there's often this box like, have you ever been convicted of a crime before? Something like that. And it's this ban the box policy. And so there's this idea that they want to remove these boxes from applications. And you've talked all about this. And so I'll let you, you know, share more about that. But when you're presenting some of your findings about what worked, it seems like it was met

16:25with some vitriol or some pushback, negative feedback, et cetera. And I love how you wrote this. And they're like, I look back at that moment. I realized I had brought data to a knife fight. And so I'm thinking about this because I'm thinking about like what are some of your reflections or thoughts about the barriers to both understanding and implementing changes that actually work, changes that align to our good intentions. And some of this might be navigating politics or otherwise. So I think we can start with the barriers and then later move into what actually works and what you've learned about that. So this is my first real experience as a professor engaging in the policy conversation. It was the first paper I had that like really kind of got a lot of policy attention and was part of an active policy conversation where the evidence was really relevant. Or, people should have viewed it as relevant. And so what I found and it was my paper and then there's another really high profile paper that came out about the same time. And both of our papers found

17:30that what happens when so the end of the box removes that question, that box that you have to check to saying if you have a criminal record. And the idea is this should help people get their foot in the door. Now anecdotally, employers just throw away your application. If you check that box, that's of course incredibly unfair and demoralizing to people who have a criminal record and have done the work and now would be great employees. So how do we fix this? And I think our impulse as human beings is whenever someone is using information in a way that we wish they didn't care about, we just remove the information. Let's just see the information away and then they can't use it and then everything's great. And I as an economist looked at a policy that removed this question from the job applications and thought, oh, like what might happen here is we haven't changed anything about the incentives employers are facing. They cared about this information for some reason. They wanted it and they were using it. And now if we're telling them they can't ask if someone has a criminal record, I worry as an economist based on everything we know from theory and other

18:30empirical evidence that they're not just going to forget about a criminal record. They're going to try to guess who has a criminal record and avoid people that they think might have a criminal record and given big racial disparities in the United States and who has a criminal history, what they might start doing is using race as a proxy and then increase racial discrimination. And so, I have a study, there's this other really cool field experiment kind of testing this in different ways and both studies found yes indeed that is what happens when ban the box goes into effect, you see big racial gaps increase in terms of who's employed and who gets a callback or an interview for a job. And in particular, the people who are losing out are young black men who don't have a record because they can't signal that out front and then employers are assuming they have a record. So, this is not good, right? It's not helping the people we tried to help. It's hurting people who are already struggling in the labor market for a variety of reasons that we should be worried about. There's a big cost here and you know, the study came out. I was at Brookings at the time, it's on sabbaticals, it's visiting fellow and so wrote some of our beds and then was asked to testify before Congress at some point as part of this whole conversation. And I wound up being on

19:34kind of one side and the person on the other side, the ban the box advocates, instead of engaging with the evidence, which is of course, is what any academic would expect. You know, if you don't like my results, like tell me where you think I went wrong, right? The advocate that was kind of brought in by the other side, quote unquote, just attacked me personally, just said, you know, she doesn't have the lived experience to understand this. She doesn't care about people with criminal records, she wishes they would all just like disappear. I think you use more colorful language than that. And I was sitting there like, wait, can we get back to like, where do you think I did wrong here? And this, you know, I, and also what I tried to make clear in my testimony, but in retrospect, this was, you know, this was never going to be a conversation. Congressional testimony is a performance. It's not a conversation. When I tried to make clear in my testimony, and I think I still try to do this is just to make it really clear, we have the same goal, right? Like we both want to help people with criminal records get jobs. That's really important. I share your concern

20:35about this problem. The question now is how to do it? Like what's the most effective policy to do this? And, you know, the evidence was sort of piling up against ban the box as a way to do this. And unfortunately, there was still a really strong political lobby that in my view, it kind of lost side of the goal and was focused on just implementing the policy as their end goal. So yeah, that was the moment I felt very ineffective in that. But now that I sent a lot more time in the policy space, I do see, you know, these congressional hearings really are more of a performance and a chance for everyone to get up and say their piece, but no one's actually listening to each other. The real conversations happen behind closed doors. So being able to be a resource to policymakers in those more closed door settings, I think is much more effective. But I think also I probably could have done even more to double down on. We have the same goal. And here, you know, 10 other things we should be trying instead, just to make it really clear, like I'm not against solving this problem. I just, I think this policy is a distraction. I'm so curious. And maybe it's not this particular

21:39example, but I remember you also writing in there as something like, and again, correct anything that's wrong here, but like unfortunate that I have thick skin or something like that. I can kind of roll with the punches type thing like a resilient sort of thing. And I've been exploring this question. I'll offer just a little bit of context and framing and then kind of bring it back and ask you about your own experience with this. But I think about that idea of having thick skin or being resilient and continuing in the face of some of these challenges that you've described, or so easily can fall off or close off or whatever it might be. And there was a guest named Mark Nipo, who's a spiritual teacher and a poet, probably one of the great poets of our time. And he said during conversation in 2018, he asked this question, which I've been exploring with incredible guests over the years or the past eight years. The question was, when my heart begins to close, how do I open it again? And the way that he shared this question was that it's not like a simple

22:41question with a static answer, but may best be viewed as a question we sort of carry with us and live into the answers over time in our own life. So I've had people like who are constructive dialogue experts come on the show and talk about when you're having a serious conversation and about trying to bridge a divide for lack of better metaphor, you can easily start to close off to the other person in conversation. We all sort of know this experience. Well, Mark's question was, well, how do you open back up again so that you can be in relationship so that you can even if you disagree, how do you stay open and maybe curious and listen and whatever it might be? So I'm so curious for you, just in your own experience, like how you've been able to open back up again and continue the conversation, because here you are in this one example, presenting data, presenting your case, and this other person gets up and it's like personal attacks. And I

23:42don't know about you, but for me, when somebody is attacking my character, it's like so quick to close off and get defensive, right? So how do you think about that? Like how do you relate to that question in the context of your work? Yeah, I mean, to be clear, I definitely don't have it all figured out. I'm not sure I'm being that individual again. There is an element of just like, okay, you're not operating good faith. It's not worth it. Like I, or maybe a better person would have tried to actively build a relationship with that person and find a way forward. I think also by nature, I think partly I'm just a contrarian by nature. I think and so I really bristle at this, you know, red team blue team, you're either with us or against us, kind of overall vibe. Oh, we have going on in the country and probably around the world. And, and so try very hard to be very difficult to pin down as being on, you know, one side or the other. And my goal really is, like, I will talk to anyone who wants to know what the evidence is, right? If you're left right, far left far right,

24:45if you want to really in good faith engage with the evidence and help solve problems or in want help solving problems, I am there. I'm happy to talk to anybody. And I think, you know, luckily, I've had so many conversations, most of the conversations I have with policymakers and practitioners, they are there for that conversation, right? It really is kind of a, I do think it's like the fringe right and fringe left want to kind of stir up this, you know, villain hero trope. And so I find it really motivating to kind of lean into the conversations where people are operating in good faith and do just want to try to solve problems in a pragmatic way. You know, most of that frankly is at the state level in the policy space I work in, partly that is because criminal justice policy is mostly determined by the states and not the federal government. But partly is because, you know, state lawmakers have to balance their budget every year. So they have to compromise on some things. Federal lawmakers don't have to do that, turns out. And I think they're just a little, state lawmakers just a little more removed from the cable news at all. And so, you know, getting

25:50some quick sound bites is just less important. And so that just makes for better conversations. And so that becomes my, like, I just, I keep my eye on the ball of solving problems, I think. And then it's like, if someone is not interested in really engaging to do that, I move on and find somebody else. And luckily, there are plenty of people out there who want to have that conversation. So one other thing that I'd love to share with you and just get your take on it from your experience. I interviewed this researcher named Kurt Gray. And I don't know if you've ever heard of him, but he wrote a book called Outrage. And he studied, he actually studies this, like in his research lab, at all, I think he's at Ohio State, like, how do we bridge the political divide and what works. And one of the takeaways from that book that was really interesting is he leans heavily towards data. He's very interested in data. But in trying to actually build a bridge with, quote, unquote,

26:51the other side to move constructively forward, he said data is not so convincing to people. And it was sort of surprising for him because he's like the facts are sort of like the data is right here. Like it's fairly obvious when you're like, you know, you can point out anything that you'd like to where we did my wrong or whatever it might be, but it's right here. But you know, like in this age where people are going, oh, well, those are your facts. And these are my, you know, so we talked about, well, how do you get around that? And one of the things that he pointed to, which for him at the time, anyway, he said was surprising, was just the power of like storytelling. And sharing stories. And a reminder of these other people I've talked to who are in the space of constructive dialogue. And so many of them talk about when we're trying to tackle really hard issues, like you're talking about in your book and in your work that you're trying to tackle as well, the power of storytelling to get people to integrate the policy or to sway opinion or to influence can be really helpful. So I'm curious if like that lands with you or if that's a mystery of

27:54a different experience. Yeah. So I join Arnold Ventures, which is a philanthropy focus on evidence based policies. They're sort of, they have research funding, but they also fund policy and advocacy to try to be the lobby for evidence based policy. I joined AB about three years ago. And I started the same day as my colleague and now friend Kevin Ring who led the criminal justice advocacy team. And we had lots of conversations on this early on because he's of course like, oh shoot, this researcher is coming in. And she's going to make me go out and tell all these lawmakers like, we don't know, we don't have any answers, right? Here's a white paper, right? And he's like, that's not what sells anything. And I think, you know, we quickly landed on a deal where, you know, I as the research director could sign off as something is being evidence based, you know, this is, we have the evidence we need that gives us more confidence that this policy will succeed if we implemented somewhere else, which is why we have research. And then he can go sell it however can use the sell it, right? And you're right, it's often storytelling, it's bringing in individuals with amazing personal stories. It's trying to relate to the lawmakers on issues they care about

28:58because they have some personal connection to the issue. So I totally recognize that that is the way that people are convinced over time. But the other thing I've learned from Kevin is that in general, you learn from people you love. And so the other thing that has come up a lot, and then we spend a lot of time on over the last three years is just bringing people together using our convening power as a funder. So we could bring people together, I would do this on the research side, we had this bridge event series where we had 12 researchers, 12 practitioners coming together for a day to brainstorm on some talk. Awesome. And so we found is like, you know, they could like meet each other, break down the, you know, silo networks or the, you know, the barriers between those silos. And they can build trust a bit because often these two groups don't necessarily trust each other. They've had bad experiences with the other side at some point. And on the policy maker side, we would bring researchers and state lawmakers together along with the AV team once a year. And, you know, there are all these channels we present them with lots of information.

30:00Our goal was not necessarily for them to leave that event with a bunch of new stuff memorized, right? With like all the white paper links and like in their head, the goal was for them to walk away thinking, those are no dentures people seemed really smart and cool. And I like hanging out with them. And then they'll call us later, they trust us and they'll call us later when they do have a question and they'll trust us to be objective and point them in the right direction. And so, you know, over time, then once you kind of meet people and get to know them as people, then you can have conversations about the evidence. Like this would have conversations about anything else. And I think then it can be persuasive. But absolutely, I completely agree with the heavens of the world that just handing someone a white paper is not going to change any one's hearts and minds. I love that. There's this quote by Bernabé researcher, Bernabé Brown. It's hard to hate from close up. Move. Yeah. Yeah. And I just love that. What you mentioned about building trust and building relationship and getting to know people as people. And then yeah, you can really have more constructive conversations.

31:02Right. Like you just once you're actually in that room, it was just amazing to see, especially all these lawmakers from across the political spectrum, far left to far right. All in the same room because they care about the same issue and they're trying to, you know, do something with domestic violence or they're trying to do something about all the money we're spending on presents, right? Or whatever their challenges in their state. And kind of realizing like, wait, I might never have a conversation with you if you are, you know, in the same legislature as me because you'd be on the other side, but it turns out we both are trying to solve the same problem in good faith. And so how can we work together to do that? And there were so many examples of these bipartisan efforts that would come out of that event for exactly that reason. It kind of gave people a chance behind closed doors to realize they actually had more in common than they thought. And that was just so cool to see. Oh, it's so cool to see. It's so needed. And it's hopeful, you know, you're like, I'm getting us into hope as I listen to you talking about this. Yeah. Oh, yes. Yeah. So we've covered, you know, maybe a couple barriers. I'm curious if there's anything else you want to highlight in this category of like what gets in the way of pushing forward effective change.

32:08Yeah. I mean, the other the other big thing we haven't talked about is just like what good research looks like, right? I think a lot of people in the policy space here or someone like me say, we need rigorous impact evaluations or strong causal research and economics language. And they think, great, so you will need a randomized control trial for everything or it's worthless. And randomized control trials are of course like that's the the gold standard. That's what we expect in clinical trials for drugs and everything else. And I think it's important to realize like sometimes that's possible. And we should be trying that more often. But it turns out like the main skill that economists bring to the table is that we have developed an empirical toolkit to use what we call natural experiments or context out in the real world that sort people into treatment and control groups as if it random as if we were in a lab. And we're very good at kind of creatively finding those situations in the real world to be able to give us traction on the underlying question of did this program or policy actually add value or not? Then nerd in me, by the way, as I was reading

33:09this because you would do this again and again in the book. And I was really brilliantly done. I was just like, this is so great. It's so cool. I you described all these different things. But maybe like maybe there's an example of one. Sure. Like you could kind of walk through so people can get a sense of like what this is. What they look like. Yeah. So so one really common design in the space and empirical design in the space is an example of where a randomized trial isn't possible is thinking about like what's the impact of prison on someone's outcomes, right? And you might look at that and say, well, we can't randomly assign people to go to prison or not or what the different sentences are. Nor should we, right? That would be illegal. I'd be an unethical. We obviously cannot run an RCT there. But it turns out in practice, we do routinely randomly assign different cases to different judges. And in practice, different judges, because their human beings have different preferences, have different views about what kind of sentence might be appropriate and different kinds of cases. And so in practice, what's what this means is that randomizing cases

34:10different judges effectively randomizes at least a subset of those cases to different treatments, longer sentences or shorter sentences, depending on whether you got lucky and got a lenient judge or unlucky and got a harsh judge. And so that's the sort of context that someone like me as an economics researcher can run with to see, okay, I can use that to measure what's the impact of having a longer sentence on your future outcomes, on recidivism, on employment, on your family dynamics, whatever it is that we might care about. And so, yeah, it turns out those kinds of natural experiments are just all over the place. And that is the fun part for someone like me, like, finding it's almost like, I don't know, it's almost like a scavenger hunt, like trying to find those in the real world. Like that's the fun part of any economics research stock. So yeah, I try to showcase in the book just how someone like me finds those natural experiments and how they work because I think there's just such a misconception out in the policy world about what research requires and what we have to do in order to test whether something works. And it's just it's much easier than

35:13I think most people realize and we could be doing a lot more and learning a lot more faster if we knew that this was possible. You know, I just read a sub-stack post that you shared with me about the work that you're going to be doing next. But you talked about kind of what you and the team have accomplished together at Arnold Ventures over the past few years. And I feel like one of the kind of key takeaways, which is tremendous, has to do with this focus on like long prison sentences versus, you know, like, and maybe you could talk through that a little bit because I think a common belief around that is maybe like longer might be more preventative of crime in the future or whatever. So there's like a story that's kind of in the common ethos about what longer means and that it might be a beneficial thing in some ways. So I don't know if you could share a little bit about that. Sure. Yeah. So the basic idea is that we, you know, if we're worried about crime, we want to deter and prevent crime, threatening someone with a really long prison sentence, making prison sentences longer should accomplish that goal, right? It should be a highly effective

36:16way to accomplish that goal partly because it keeps dangerous people off the streets if they're behind bars and partly because the threat of a long sentence should deter people from committing crime. You don't you want to avoid that and so you stay out of trouble. And honestly, like that's exactly the kind of policy that an economist might design, right? It's sort of based on this idea of people responding to incentives. It all sounds great. But we've tried this now for many decades and we've learned a lot as we've gone and it turns out it doesn't work as well as we had thought for two reasons. One is that most people who are at risk of committing crime, people were trying to deter simply aren't thinking that far ahead. They're not thinking past today or this week. And so making a sentence two years instead of one year or five years instead of two years or 10 years or 20 years, you know, and some of these sentences are really long. None of that is registering today when they're making decisions about how they spend their time or whether to go out with those friends who are always getting them into trouble. What does register, what does change behavior

37:17reliably is the swiftness and certainty of consequences. So if you know that with a very high probability, you're going to get caught if you rob this 7-11, then you're going to be much less likely to do it. And we see that in lots of different scenarios. We have lots of strong empirical evidence from these kinds of natural experiments I'm talking about. The other reason long prison sentences don't work as well as we thought is that for most people, and of course there are some exceptions that prove to the world that for most people crime is a young man's game. And so you age out of crime really quickly. And so locking someone up as someone commits a crime when they're 18. By the time they're 25 or 30, it's as if they're a completely different person. And we can all think about how we have matured and maybe make better decisions and take fewer risks than are less impulsive than we were when we were a teenager. It turns out everyone, the vast majority of people are the same way. And so locking someone up when they're 18 until they're 25, that might prevent a lot of crime. Locking them up until they're 40 or 50 or 60, there are a lot of prisons that are now like essentially operate nursing

38:19homes, hospice, carer, wings, and that's, you know, at that point we are getting zero public safety value from that. It's entirely about retribution at that point. And so I just want us to be honest about what we're paying for. And so what we try, what my team has been really pushing instead, because of course if you want to do something about crime, okay, long prison sentences are not adding the value we thought they would be. So what should we do instead? It turns out that increasing the probability that you get caught. That's the real lever we should be pulling. So what does that mean in practice? Solve more crimes faster. And so we've been spending a lot of time going around to police departments and states and trying to figure out how do we get more resources to law enforcement to help beef up their investigation resources, right? How do we help? How do we help them hire more detectives or train detectives in a different way or reduce the backlogs in their crime labs or help them buy rapid DNA machines, things they can help them solve more crimes faster, because we have all of this empirical evidence at this point that that will have a huge deterrent effect on crime. And then that pace is double-divine because the crime never happens

39:22and you don't have to lock anybody up. And so it's just a win-win, right? It turns out it's way cheaper and more effective. And so, you know, we should be doing more of it. Can we spend a little bit of time here because I realized we're, you know, sort of moving toward a close in the conversation, but maybe along these lines of what works and sort of win-wins? I mean, I can think of a number of them, but I'm curious, like, what stands out to you as maybe something that was really surprising or particularly beneficial or something that you just feel like would be worth highlighting? Yeah, so that's a big one, right? So all more crimes faster, along similar lines, some of my own researchers on the effects of DNA databases. So these are used now around the world. Every state in the US has a DNA database that are they're all linked through the FBI so you can search across states. And the idea is that if you're added to the DNA database, and it's just a string of numbers, it's kind of like hashed so security number, it's not

40:24your whole genome. So less of a privacy invasion than it might sound like you're more likely to get matched to DNA from crime scene evidence from unsolved crimes. And so you're more likely to get caught if you refund. And I have research paper from Denmark that has amazing data on this, so we're able to really dig in deeper than we could in other contexts. And comparing people who are charged just on either side of the implementation date of the law, one person gets lucky and they don't go on the database, the person that charged the next day gets unlucky, they do go on the database, you can compare those two people over time. And the person who just by the luck of the draw, went in the database is 40% less likely to reoffend for zero. It's just, you know, game changer, right? And so this is just the order of magnitude that we can think about if we can increase the probability of getting caught. So that's one. Another type of evidence like it completely unrelated way outside the box that I'm still blown away by is thinking about the impacts of the real time impacts of exposure to air pollution on violent crime. And so I think we're used to thinking about

41:28exposure to pollution or even to lead when you're a young child that affects your brain development, when you grow up, you're more likely to be violent and impulsive and wind up in prison. That certainly happens. But the evidence that's accumulating on how exposure to pollution air pollution in real time affects the way our brains work makes us less productive on the job, it makes us more irritable, turns out also increases violence is still really striking to me. My favorite study on this is, you know, looking at a highway that runs through Chicago. So there's exhaust from the cars on the highway. And some days the wind blows that exhaust to the west and you see violent crime go up in the west. And then the next day it blows the exhaust to the east and you see violent crime go up in the east. So the violent crime is like following the wind direction, which the only explanation is that it's the air pollution that's doing this, right? And so man, like why don't we try putting better air filters into schools and jails and public housing and other public buildings, right? Like me, that could really help in a way that's super cheap.

42:30So that's one that again, I still find really surprising and the evidence is really piling up that this is a real thing. Is this like just set of curiosity? Is this long term like ongoing air pollution or is it more like, so what I'm thinking about right now is I just was talking some of yesterday and the wildfires are burning in Ontario and Canada and the smoke and the air and the quality of the air like the air index is like, I think it was the worst in the world where they were. Yeah, yeah, I just wrote a subset post about this this morning because yeah, seeing all that photos are coming out of this, it's really crazy. So yeah, we would expect based on all the research, including some specifically on the effects of wildfire smoke, is that we should expect violent behavior will go up in real time as people look at those to this. So anywhere where, you know, if you're in a jail or prison that doesn't have air conditioning, so it's not going through an HVAC system, right? You're exposed to that air pollution. I would expect more violent incidents in those facilities, schools, if they don't have a great air filtration system,

43:30I would expect more violent behavior or more disciplinary infractions of some sort. In general, the way to think about it is it just pushes everyone a little bit closer to snapping, like, losing their temper, right? And there's also something about like the underlying biology here is also partly it makes it harder for you to gauge the other person's kind of emotional state. And so it's harder for you to read emotional cues. And so that might lead to more misunderstandings that it might, might normally. So just a whole bunch of different ways that this seems to be affecting the way our brains are operating in ways that push us toward more violent confrontations. So yeah, air pollution, it turns out people think of it as entirely a public health issue. It's also a public safety issue. And here's another one that it kind of relates to that idea of like people getting closer to snapping, right? It was this, and I don't remember if it's Norway. So maybe it wasn't Norway, not sure. But there was this prison example you gave where it was much more open and freer like their doors run locked and they could go out and they had jobs and

44:34all kinds of things. So that was a really like just eye opening case study for me. And you had many of these in the books, but could you talk a little bit about that? Sure. Yeah. So I think Scandinavian prisons like Norway are sort of the typical example of this. And they get the most attention among the prison reform crowd. But I actually find examples from other countries a bit more compelling and interesting because there's so many differences between the US and Norway. It's like, you know, taking any one piece of their system and moving it over how confident can we be that it's we're going to get the same result. So Italy has this a couple of examples, what they call open prisons relative to the standard closed prisons where you can move around more easily. You're there for a long time. You can come and go from the prison and have a job, a bunch of different things like that. And the underlying hypothesis is that more responsibility begets rehabilitation. So there are now some ongoing randomized trials in the United States in a few different prison systems in Pennsylvania and Missouri in South Carolina where people are being

45:35at random sorted into normal prison units or more Scandinavian style prison units where you have more openness. Another favorite example of this sort of innovation and experimentation is the Netflix show unlocked. Which is if you haven't watched this is worth watching. It's sort of the first time I watched the first episode I was like, wait, wait, wait, this can't be a documentary. It's can't be a real thing. How they get the cameras in there. How do they do this? But it is. And so basically the idea is and this is pushing a lot of innovation around the country. Jails and prisons are across the United States are incredibly understaffed and they're incredibly violent because of the understaffing. And of course that leads to just just like a negative spiral, right? It's because it's so violent, more people quit and you have even fewer staff. And so Jails and prisons across the country are really being forced to say like, okay, we've got to think way outside the box here. What else can we try? And so these Jails that are documented in this unlocked series say, okay, we're going to try pulling all the guards out of the unit. We're going to unlock all the doors and you all have to

46:36be a community and figure out how to run this yourselves and take care of each other. And there are ups and downs for sure, but in the end, it winds up working. And so it's just fascinating to watch, you know, necessity, be getting an engine, I guess, in these different contexts where we might normally think people are really reluctant to try new things. You know, one of the things along these lines that I really appreciated too, just stood out to me in the book was there was this woman who came in and I think she was leading an organization, I remember correctly, I think her name was Susan, but you can correct that if it's not. And she basically came to you and your team, I believe, and said like something like, I just want to know what works. I want to know that the impact that I'm having or doing through my work actually is having a positive impact. And so I really want to understand that and you spent some time telling her story a little bit and talking about the courage that it takes to actually open ourselves up to whatever is really true

47:37about what we're doing in the world. Because, you know, kind of this bias towards like, oh, well, people tell me I'm doing a good job and I think I'm well intentioned and maybe the data plays out that way, but maybe it doesn't. The worst fear being, wow, my good intentions are actually doing the opposite. They're actually harming the people that I'm trying to help. And you know, I think it's so easy for someone like me on the research side to be like, why aren't we just running more experiments? What's wrong with me? You know, like, doesn't just not want to know, you know, and putting, putting yourself into the shoes of someone who has run an organization, it's been their baby for like maybe their whole lives. And, and it is a risk to engage in research like this because what if it shows you're not actually having the big benefits you thought you were, right? That, you know, kind of feels like there's only downside risk then. If you currently believe it's really helping. So yeah, this was actually at an amazing convening that the lab for economic opportunities or Leo held into research organization at the University of Notre Dame. And they do a lot

48:38of work with nonprofits and government agencies to help them do these rigorous evaluations of some program or intervention they're trying. And there was a woman there who was leading one of these nonprofits they work with. And she stood up and said, you know, I went to the retirement party of one of my mentors, someone who ran this organization for a long time. And he confessed to them all in his kind of going away speech, right? That he wakes up some days and wonders if he had any impact at all on any any of the lives of any of the people who came through that that nonprofit served. And she like, you know, and standing in front of this whole convening hundreds of people in this room and said, in that moment, I realized I never want to wonder. I don't want to get to the seems like the worst case scenario was I get to the end of my career and I look back and I am not sure if I had any impact at all. And so she found Leo and it's been working with Leo and wants to do the rigorous evaluations. And she's like, look, if what I'm doing is not having an impact, I want to know it so that I can do better. You're like, I owe it to the community I'm trying to serve to,

49:38like, I want to know at the end of the day, I did everything I could to maximize my impact on their lives. And so she viewed it as having a personal responsibility to engage in this work, even though it was, you know, it was really uncomfortable a lot at the time. And it was opening herself up to no results, right? Or maybe even, you know, results showing you're maybe even doing harm, like in that worst case scenario. But she's like, I want to know if that's the case and I want to fix it. And it just struck me in that moment, how, how much courage it takes for people on that side to engage in this work that I take for granted is obviously good. And I think, you know, I'm now a spent a bunch of time in this role at AD, talking to policymakers, talking to practitioners, trying to pull them toward doing more rigorous evaluations. And it's actually been amazing by the, you know, just wrapped up my three years and I'm moving on now. But it is really, it's been amazing to me that at this point, the biggest constraint on having more

50:41and better research is not that the folks on the policy side don't want to engage in it. It's that we don't have enough researchers. Like we, we've been able to get people are open to it and interested in like willing to lean in. And we don't have enough people to match them with to do the studies. And so, you know, my next chapter, I'm going to be thinking a lot about how to build up that research pipeline so we can meet the demand. But it's been really, I mean, especially with that story. And I called her season in the book. I actually don't, I don't think I ever actually knew her, her, you just a participant in this conference. But that's, I found that story, her story that she told just incredibly moving. I am just so grateful for the amazing work that you're doing in the world. Thank you for writing this book and putting it out there. And I'm curious if there's some sort of closing thought, something you'd like to leave people with, it could be around something we explored earlier in the conversation today that you want to underscore, or maybe it's something new we haven't covered yet. It's coming to mind for you. Yeah. Well, I mean, I just mentioned Briefly I'm moving on from AV and part of this is honestly just because I feel so motivated and so optimistic

51:44about the potential to make incredible change in this space. And, you know, so I've been working for this philanthropy for the last three years. Basically, our job was to give away Laura and John Arnold's money. They're putting toward this. They could be sending all of their money on their billions on yachts and islands and they're choosing to spend it on research and evidence-based policy lobbying. And I am just so grateful for them to do that. And also, it's crazy that they are basically the only funder of this kind of working the United States. And so I really want to try to bring other donors into this space so that we can go bigger and faster, put more boots on the ground. It became so clear. As I just said, you know, policymakers and practitioners are eager to do these evaluations. We don't want them into researchers. Okay, let's build up that research pipeline. And then on the policy side, state lawmakers are very open to all of these ideas and this evidence. And they just didn't know about it. They work part-time. They don't have a staff. They just genuinely did not know that solving crimes faster works better in the longer sentences, right? So just trying to get the word out about what we're learning turns out it can move policy pretty quickly when we're

52:45able to have more of those conversations. So I am just, you know, in a broader space where I think people are very cynical these days about whether data and evidence matters anymore and whether, you know, the two sides can work together. I just see so many examples every day of both of those things continuing to matter and, you know, people working together across the aisle and caring about evidence and wanting to solve problems. And my next chapter is going to be to try to do even more of that. Thank you so much, Jennifer. It's like a joy to be with you and I'm so inspired by your work. And it, you know, makes me more hopeful. So thank you again. If people want to find out more about what you're doing next, keep in touch with you, follow your writing or podcast or the book. Where's the best places for them to do that? Yeah. So I am at Jennifer Dulliac on X. I'm on LinkedIn and I have a sub-stack. It's called probable causation, which is also the name of my podcast, which has been running for gosh, I think like seven years now, something like that insane. So yeah, and I'll be

53:48writing kind of some reflections on how philanthropy in particular can be catalytic in this space and what I'm thinking about and evidence-based ways to improve public safety. Thanks for listening to another conversation on the courageous life. Every episode we put out is made possible by the extraordinary partnership with executive producer, professional musician and composer, Matthew Donor. For more on his incredible work, original music and photography, please visit MatthewPatrickDonor.com. And for more on the courageous life, my live events and offerings, check out practicing courage.com. Thanks again for tuning in. Until next time, this is Joshua Steinfeld with the courageous life podcast.